Mammalian genes can be grouped into five isochores based on their GC content, with genes of L-isochores (low GC) uniformly expressed at a lower level than the genes of H- isochores (high GC). Our preliminary data strongly suggest that influenza genes, similarly to genes of other viruses, are uniformly biased towards L-isochores and thus, are likely to be under-expressed in vivo, both during natural infection and especially if introduced in the context of a recombinant vector. We propose to increase the level of influenza viral genes' expression in a standard DNA plasmid vector by altering their codon usage from L1- to H3-isochore (the latter being most highly expressed). We propose to modify according to this algorithm, four conserved genes of influenza virus that are known to possess strong potential for induction of cross-protective immunity (NP, M1, NS1 or M2). We previously demonstrated limited, yet statistically significant protection against heterotypic influenza viral challenge in mice and chicken vaccinated with a combination of plasmids encoding NP, M1 and NS1 genes. All of these genes are naturally biased towards L1-isochore. We will substitute wild-type codons that possess such an L1-bias (16-20% of total coding sequence) with synonymous ones that will adhere to the codon preference of the highly expressed H3-isochore. We will then compare the expression levels of modified conserved influenza genes to their wild-type counterparts in vitro. If the expression level is enhanced, we will propose to utilize the modified genes to increase the immunogenicity and protective efficacy of the DNA vaccine during Phase II of this grant. We have already developed an in silico technology for synonymous codon changes, that allows the generation of gene modification that adheres to the codon usage of a specific isochore family. It is known that the expression of many viral proteins in various vector systems is suboptimal. This phenomenon profoundly hinders the development of recombinant vaccines against different viruses including influenza. Accepted means of codon modification (i.e., humanization ) of viral sequences do not necessarily lead to increased expression, probably because isochore adaptation has never been taken into account. Therefore, if proven, our approach will lead to a novel paradigm for recombinant gene modifications for enhanced expression in vivo, both for vaccine and non-vaccine applications. [unreadable] [unreadable] [unreadable]